Abstract: Scanning probe techniques are popular, non-destructive ways to visualize the real space structure of Van der Waals moirés. The high lateral spatial resolution provided by these techniques enables extracting the moiré lattice vectors from a scanning probe image. We have found that the extracted values, while precise, are not necessarily accurate. Scan-to-scan variations in the behavior of the piezos which drive the scanning probe, and thermally-driven slow relative drift between probe and sample, produce systematic errors in the extraction of lattice vectors. In this Letter, we identify the errors and provide a protocol to correct for them. Applying this protocol to an ensemble of ten successive scans of near-magic-angle twisted bilayer graphene, we are able to reduce our errors in extracting lattice vectors to less than 1%. This translates to extracting twist angles with a statistical uncertainty less than 0.001° and uniaxial heterostrain with uncertainty on the order of 0.002%.
Introduction
In the field of 2D materials, "twistronics" has emerged as a dominant paradigm. When two layers of graphene are stacked with a slight relative twist, they form a moiré superlattice. These structures can host exotic electronic states, including superconductivity and correlated insulation, which are exquisitely sensitive to the exact twist angle and the presence of "heterostrain"—minute relative deformations between the layers.
While scanning probe techniques like Atomic Force Microscopy (AFM) are standard for visualizing these moiré patterns, they often encounter a fundamental discrepancy between precision and accuracy. Systematic distortions arise from the non-linear behavior of piezo-actuators and thermal drift during the scanning process. These effects warp the resulting images, leading to errors in calculated structural parameters that can exceed the physical variations being investigated.
Methodology: The Correction Protocol
To address these systematic distortions, we developed a two-step post-processing protocol designed to "unwarp" scanning probe data:
- Scan Area Correction: Rather than assuming a perfect scan grid, we utilize raw data from X/Y position sensors. This allows us to map each pixel to its true physical coordinate, correcting for piezo non-linearities and scaling errors.
- Drift Correction: Thermal drift acts as a constant velocity shift, analogous to a Doppler shift, on the observed lattice periodicity. By performing successive scans with alternating "slow-scan" directions (Forward vs. Backward), we can isolate the drift velocity and mathematically transform the distorted lattice vectors back to their intrinsic values.
We validated this approach using Torsional Force Microscopy (TFM) on near-magic-angle twisted bilayer graphene (TBG). TFM is particularly effective for this application as it is highly sensitive to the atomic-scale friction variations characteristic of moiré patterns. By moving to "open-loop" scanning and correcting via sensors afterward, we can maintain high accuracy even at faster scan rates.
Key Results and Discussion
Applying the correction protocol to ensembles of successive scans significantly improved the accuracy of our structural characterization. We reduced errors in extracting moiré lattice vectors to less than 1%. This high level of accuracy allowed for the determination of the twist angle and uniaxial heterostrain with unprecedented statistical certainty.
Specifically, we measured a twist angle of approximately 1.126° with an uncertainty of less than 0.001°, and uniaxial heterostrain with an uncertainty of ~0.002%. These results remained consistent across different scan speeds (4 Hz and 8 Hz) and scan directions, confirming the robustness of the method. These uncertainties represent an order of magnitude improvement over typical values obtained from uncorrected microscopy or electronic transport measurements.
Conclusion and Future Outlook
While the current protocol assumes constant drift and uniaxial strain, it provides a powerful framework for high-precision structural mapping of Van der Waals heterostructures. Future work may extend this model to account for non-constant drift or biaxial strain components. Ultimately, integrating this high-accuracy feedback into automated "robotic stacking" systems could enable the creation of perfectly aligned moiré devices with predictable and reproducible electronic properties.
Summary updated